Super Bowl LIV DVOA/Quick Reads
The Super Bowl ended up as another one of those games that happens sometimes in the DVOA system, where the team that's more efficient over the course of the game loses because it groups its inefficient plays into three-and-outs (and in this case, four-and-outs) and thus ends up running fewer plays than the other team.
San Francisco ended the game with 44% DVOA, while Kansas City ended the game with 21% DVOA. This suggests that based on the play in this game, we would expect San Francisco to be the better team in the long run. But of course, the Super Bowl isn't about the long run. It's just one game, and there's no tomorrow.
San Francisco outgained Kansas City on average, 6.5 yards per play to 5.3 yards per play. That latter number is a bit deceiving because of the big losses Patrick Mahomes took when kneeling out the clock at the end; take out those kneels and his final time-wasting incomplete pass, and the number is a bit closer, with the Chiefs averaging 5.8 yards per play. Both teams had two interceptions. The Chiefs did fumble three times but maintained possession on all of them; the 49ers fumbled only once, also maintaining possession. But Kansas City ran 75 plays, 71 of which were meaningful, while San Francisco ran only 54 plays, 53 of which were meaningful (they kneeled out the clock in the first half).
| DVOA (with opponent adjustments) | ||||
| TEAM | TOT | OFF | DEF | ST |
| SF | 44% | 22% | -23% | -1% |
| KC | 21% | 16% | 2% | 7% |
| VOA (no opponent adjustments) | ||||
| TEAM | TOT | OFF | DEF | ST |
| SF | 22% | 16% | -7% | -1% |
| KC | -11% | -5% | 13% | 7% |
Oh boy, did things turn in this game, though. The Chiefs were a bit lucky to have things tied after the first half, as the 49ers were significantly more efficient, gaining 8.1 yards per play to just 4.7 yards per play for Kansas City. The Chiefs kept the game tied thanks in large part to Garoppolo's first interception.
For the third quarter through Kansas City's first drive of the fourth quarter and the interception over the middle by Tarvarius Moore, it was all 49ers. And then after that, the fourth quarter was all Chiefs.
| Offensive DVOA of Super Bowl LIV | ||
| SF | KC | |
| First half | 36.7% | 3.1% |
| Halftime through 12:05 of Q4 | 87.7% | -80.4% |
| Last 12 minutes of Q4 | -67.3% | 119.7% |
Thanks again to everyone for a great 17th season at Football Outsiders. We'll start our offseason coverage soon along with announcing the winners of the Football Outsiders reader awards next week.
Now let's turn it over to Vincent Verhei for some Quick Reads.
| Quarterbacks | |||||||||||
|
Rk
|
Player
|
Team
|
CP/AT
|
Yds
|
TD
|
INT
|
Sacks
|
Total
DYAR |
Pass
DYAR |
Rush
DYAR |
Opp
|
|
1.
|
Patrick Mahomes | KC |
26/42
|
286
|
2
|
2
|
4
|
66
|
53
|
13
|
SF
|
| Mahomes gains 83 DYAR due to opponent adjustments, and thank goodness today is the last time I'll have to mention that in this crazy season. His DYAR is dinged for two fumbles -- one on a sack, one on a run -- though Kansas City recovered them both. His rushing DYAR is based on his six carries for 44 yards and a touchdown, ignoring the 15 yards he lost on three kneeldowns at the end of the game (which caused a lot of havoc in Las Vegas). On third and fourth downs, he went 5-of-10 for 63 yards with as many conversions (two, including a touchdown) as interceptions ... though, to be fair, an 11th throw resulted in a DPI for 20 yards, and he also ran for a third-down conversion. Though he had success on throws to the outside, he did little damage on throws down the middle, going 6-of-9 for only 46 yards with an interception. As opposed to... | |||||||||||
|
2.
|
Jimmy Garoppolo | SF |
20/31
|
219
|
1
|
2
|
1
|
33
|
35
|
-2
|
KC
|
| Virtually all of Garoppolo's good plays were throws down the middle, where he went 11-of-15 for 136 yards with one interception. To his, right, however, he only went 6-of-10 for 42 yards with one touchdown and one interception. He threw for eight first downs down the middle, but only two to his right and three to his left (where he went 3-of-6 for 41 yards). Eleven of those 13 first downs came in the first three quarters; in the fourth, he went 3-of-11 for 36 yards with a sack and an interception. He did not have a good day on deep passes, completing more to the Chiefs (two) than to his own team (one, for 18 yards) in five throws. He was best on throws to his running backs, completing all six of his passes for 64 yards and four first downs, including a touchdown. And speaking of running backs... | |||||||||||
| Qualifying Running Backs | ||||||||||||
|
Rk
|
Player
|
Team
|
Runs
|
Rush
Yds |
Rush
TD |
Rec
|
Rec
Yds |
Rec
TD |
Total
DYAR |
Rush
DYAR |
Rec
DYAR |
Opp
|
|
1.
|
Damien Williams | KC |
17
|
104
|
1
|
4/8
|
29
|
1
|
47
|
40
|
7
|
SF
|
| Only three running backs met our usual threshold for qualifying for our tables this week, so we'll discuss them here, then three others who each had at least one handoff and/or target. Williams was clearly the best of the bunch. It helps that he gained 19 DYAR due to opponent adjustments, but his numbers were also quite good on their own. He's the first player to run for more than 100 yards on fewer than 20 carries in a Super Bowl since Frank Gore did it for the 49ers against the Baltimore Ravens seven years ago. In 17 carries, Williams was stuffed just one time; he gained at least 2 yards on every other run, totaling eight first downs, including a 38-yard touchdown. Three of his catches also produced first downs. | ||||||||||||
|
2.
|
Raheem Mostert | SF |
12
|
58
|
1
|
1/1
|
2
|
0
|
2
|
5
|
-4
|
KC
|
| Mostert loses 13 DYAR due to opponent adjustments. Only one of his runs was stuffed, while four went for first downs, including gains of 11 and 17 yards. | ||||||||||||
|
3.
|
Tevin Coleman | SF |
5
|
28
|
0
|
1/1
|
3
|
0
|
-5
|
-1
|
-4
|
KC
|
| Coleman had 17 yards on one carry in the second quarter, which produced his only first down, and 11 yards on his other four runs. His one catch was a 3-yard gain on third-and-5. | ||||||||||||
| Non-Qualifying Running Backs | ||||||||||||
|
Rk
|
Player
|
Team
|
Runs
|
Rush
Yds |
Rush
TD |
Rec
|
Rec
Yds |
Rec
TD |
Total
DYAR |
Rush
DYAR |
Rec
DYAR |
Opp
|
|
1.
|
Kyle Juszczyk | SF |
0
|
0
|
0
|
3/3
|
39
|
1
|
35
|
0
|
35
|
KC
|
| By DYAR, Juszczyk was San Francisco's top receiver against Kansas City. His three targets: 15-yard touchdown on first-and-10; 14-yard gain on second-and-6; 10-yard gain on first-and-10. He did not, however, get a target in the fourth quarter. | ||||||||||||
|
2.
|
Jeff Wilson | SF |
0
|
0
|
0
|
1/1
|
20
|
0
|
14
|
0
|
14
|
KC
|
| Wilson's one target: a 20-yard gain on third-and-5 in the second quarter. | ||||||||||||
|
3.
|
Darwin Thompson | KC |
1
|
0
|
0
|
0/1
|
0
|
0
|
-15
|
-8
|
-7
|
SF
|
| Thompson's one carry went for no gain on first-and-goal from the 1; his one target was an incompletion on third-and-7. | ||||||||||||
|
Rk
|
Player
|
Team
|
Rec
|
Att
|
Yds
|
Avg
|
TD
|
Total
DYAR |
Opp
|
|
|
1.
|
Travis Kelce | KC |
6
|
6
|
43
|
7.2
|
1
|
52
|
SF
|
|
| Kelce's totals include 43 DYAR receiving, 9 DYAR rushing for his one carry for 2 yards (a third-down conversion). Only two of his catches produced first downs, but five of them counted as successful plays. He also picked up 20 yards on a DPI on third-and-10. | ||||||||||
|
2.
|
Sammy Watkins | KC |
5
|
6
|
98
|
19.6
|
0
|
41
|
SF
|
|
| Four of Watkins' catches produced first downs, including gains of 28 and 38 yards. | ||||||||||
|
3.
|
Kendrick Bourne | SF |
2
|
4
|
42
|
21.0
|
0
|
11
|
KC
|
|
| Bourne's two catches: 16-yard gain on second-and-7, 26-yard gain on third-and-8. | ||||||||||
|
4.
|
Deebo Samuel | SF |
5
|
9
|
39
|
7.8
|
0
|
9
|
KC
|
|
| Samuel's totals include -22 DYAR receiving, worst of any player this week, but also 31 DYAR rushing for his three carries for 53 yards and three first downs. Three of his catches produced first downs, but the other two both resulted in losses. | ||||||||||
|
5.
|
Emmanuel Sanders | SF |
3
|
5
|
38
|
12.7
|
0
|
6
|
KC
|
|
| Sanders' three catches: 18- and 5-yard gains on first-and-10 and a 15-yard gain on second-and-5. | ||||||||||
|
6.
|
Tyreek Hill | KC |
9
|
16
|
105
|
11.7
|
0
|
3
|
KC
|
|
| We had to expand our receivers list by one name to make sure we included the game's top receiver. Yes, Hill caught nine passes for 105 yards, but it took him 16 targets to get there. And only three of those catches produced first downs; none of the others gained more than 9 yards. | ||||||||||
| Worst Wide Receiver or Tight End by DYAR | ||||||||||
|
Rk
|
Player
|
Team
|
Rec
|
Att
|
Yds
|
Avg
|
TD
|
Total
DYAR |
Opp
|
|
|
1.
|
George Kittle | SF |
4
|
7
|
36
|
9.0
|
0
|
-4
|
KC
|
|
| Only two of Kittle's catches produced first downs, and none gained more than 12 yards. | ||||||||||
Comments
60 comments, Last at 10 Feb 2020, 1:49pm
#15 by ammek // Feb 04, 2020 - 2:09am
So has a quarterback ever won MVP with fewer DYAR than Mahomes? I don't know who else could have gotten the award; maybe Kelce or Watkins. I think it goes to show that we still underestimate the (negative) value of turnovers, and especially of fumbles that are not recovered by the defense.
Thanks for all the work you do on Quick Reads btw.
#21 by Aaron Brooks G… // Feb 04, 2020 - 8:12am
The knock on Williams would be his highlights were either from Mahomes, or were based on an extra score after the Chiefs had already taken the lead.
42 of his 104 yards came on a drive that the Chiefs only needed a first down to win. To that point, Mahomes had almost as many rushing yards and more rushing TDs.
#19 by NYCityRat // Feb 04, 2020 - 4:43am
Super Bowl 42 MVP Eli Manning: 19/34, 255 yards, 2 TD, 1 INT, 3 sacks. I can't find his DYAR -- all I can find is 4.2 DPAR from an old Fox Sports column, and based on this column:
https://www.footballoutsiders.com/dvoa-ratings/2008/dpar-dead
it looks like the conversion from pre-2008 DPAR to post-2008 DYAR is between 12 and 14.5. That also accounts for DPI's being a positive factor for a QB's passing, which Manning would not benefit from in SB 42. Manning's conversion factor would have to be 15.7 to reach Mahomes's 66 DYAR. Not inconceivable for a single game, but obviously a contender.
That also doesn't account for the fact that Manning almost fumbled (thanks to Adalius Thomas) AND threw (right to Asante Samuel) the game away on the game-winning drive.
#20 by Eddo // Feb 04, 2020 - 6:32am
Good find with Manning in Super Bowl LII, his DYAR is almost certainly worse than Mahomes's. But your last paragraph is a step too far. Mahomes (and many other QBs) also had some near-miss plays, you can't only cite that for Manning to make his game seem worse.
#42 by NYCityRat // Feb 04, 2020 - 11:55am
I didn't mean to use it as evidence. I love the Giants, and SB 42 was an incredible game, but the serendipity of the helmet catch and the casual ease of the winning TD makes people forget what a sh*t show the rest of that drive almost was. The two plays I mentioned where Eli came THIS close to losing the ball (and two other passes where the Patriots had a legit shot at picking him off, one to Harrison and another to Meriwether), the fourth down Jacobs almost didn't pick up, Eli constantly being under pressure...I thought Mahomes (and KC generally) was better in KC's comeback.
#23 by Aaron Brooks G… // Feb 04, 2020 - 8:35am
What was Montana in XXIII? Lot of yards and two TDs, but two fumbles and 3 sacks against a bad defense. Similarly Rypien in XXVI?
Elways in XXXII was super-pedestrian, even though GB had a really good defense.
Manning in XLI was pedestrian, but the Bears were stellar on defense and the game was played in a monsoon.
In pre-DVOA days, XVI might be interesting if we ever get there. Montana had a pedestrian game. Maybe Bradshaw in XIV. He had a lot of turnovers in XIII, but also four TDs. Dawson in IV was underwhelming, although still the best offensive player in a game dominated by defenses.
#30 by Eddo // Feb 04, 2020 - 9:40am
Looking at the list of all Super Bowl winning QBs that threw at least ten passes, the lowest AY/A was Ben Roethlisberger in Super Bowl XL vs. Seattle (9/21, 123 yards, 0 TD, 2 INT), but he wasn't MVP.
You cited the second worst, John Elway in XXXII vs. Green Bay (12/22, 123 yards, 0 TD, 1 INT), but he wasn't MVP (Terrell Davis was).
Patrick Mahomes (26/42, 286 yards, 2 TD, 2 INT) comes in at fifth worst, and the worst MVP. This doesn't account for his rushing plays, which were quite valuable.
Other MVPs:
Peyton Manning in XLI vs. Chicago (25/38, 247 yards, 1 TD, 1 INT) is sixth worst. Tom Brady in XXXVI vs. St. Louis (16/27, 145 yards, 1 TD, 0 INT) is ninth worst. Tom Brady in XLIX vs. Seattle (37/50, 328 yards, 4 TD, 2 INT) is twelfth worst.
#54 by mehllageman56 // Feb 05, 2020 - 7:43pm
While the 1988 Bengals' defense was a pedestrian 0.2% in DVOA, they were better at defending the pass (10th in the league at -3.8%) than the run. Not really that bad a defense, especially considering 2 of the guys in the secondary were selected for the Pro Bowl, as well as nose tackle Tim Krumrie. The 1991 Bills were slightly worse at 2.1% overall, and 1.2 % for passing DVOA, but still mediocre, not really bad. Obviously, the 2019 Niners defense is way better and gives Mahommes a big bonus that Montana and Rypien wouldn't get. Elway might be a better bet.
#9 by Pat // Feb 03, 2020 - 3:38pm
Aaron said why in the writeup, though. The 49ers had significantly fewer plays. On average they were higher value, but they had less of them because the bad plays occurred early in a few drives, killing them. Garoppolo's 2 interceptions, and then the incomplete/incomplete when up 20-17, plus the mess at the end of the first half. That's 3 drives that ended really quickly. Whereas the Chiefs only had the single 3 and out.
I mean, realistically the 49ers lost it due to only getting the field goal at the beginning of the 3rd quarter. With a touchdown there, the rest of the game pretty much plays out to a 24-24 tie and the 49ers aren't nearly pressing so hard on that last drive.
#5 by Vincent Verhei // Feb 03, 2020 - 2:57pm
He actually had six games this season with worse passing DYAR than this. His worst game this year was -17 passing DYAR in Week 17 against the Chargers -- and that was not meaningless, since KC needed a win and a NE loss to get a bye (which obviously ended up happening). Mahomes was 16-25-174-1-1 in that game. The Chiefs still scored 31 points, but they needed a Mecole Hardman kickoff return for a touchdown and a 84-yard touchdown run by Damien Williams to do it.
#6 by Vincent Verhei // Feb 03, 2020 - 3:18pm
Well, it's not that simple. To be more precise, Thompson's one target, an incompletion on third-and-7, was worth -6.6 DYAR. Take away opponent adjustments, it was worth -5.7 YAR. The average throw to a running back on third-and-7 gained 4.4 yards in 2019, so just on raw real estate Thompson was well behind the curve on an incompletion.
Then you throw in the value of a first down. Third-and-7 passes to running backs are low-percentage plays, but they're not hopeless -- they converted 26% of the time this year. Those 26% got a bonus for picking up the first down, which means the 74% of the plays that failed were penalized for not picking up first downs.
Then you throw in adjustments for score, field position, etc. on top of that.
#10 by Cythammer // Feb 03, 2020 - 7:03pm
I think it makes sense that SF ended with a better VOA. It matches what I saw while watching the game. The 49ers dominated the game for 3.5 quarters and should probably have been well ahead. It took a fourth down conversion for KC to get their one TD before the turnaround. Even once things changed, it wasn't like KC scored easily. Obviously we all remember the huge play that came on third and fifteen. Later on they faced a third and goal from the 10 On the drive that took the lead it took until third and goal before they scored and it almost went to fourth down. Overall it was an odd outcome for how the game played out.
#13 by mrh // Feb 03, 2020 - 10:35pm
I don't buy the idea that the 49ers dominated thru 3.5 quarters. At the point of the 2nd int thrown by Mahomes, the Chiefs led in ToP about 27 minutes to 21 on 53 plays to 39. That INT could easily have been a catch or incomplete, leaving the Chiefs with a FG and within 1 score. I'd say the 49ers were better - those turnovers counted - but not dominant.
The 49ers WERE more efficient, but the Chiefs had been efficient enough to have sustained a few drives.
As for getting the only TD because they went for it on 4th down, it's not illegal to make good tactical decisions. Similarly, the decision to go hurry up on the 1st TD drive in the 4th qtr was good tactics not just from a clock management standpoint, but also considering that the SF defense may have been a bit tired (the 49ers rarely lost ToP all year).
#36 by Cythammer // Feb 04, 2020 - 11:03am
I don't think having more time of possession and running more plays indicates the Chiefs were playing better. Running more plays just means they were less efficient, and pretty much the same thing for time of possession. The point of mentioning the fourth down is to show that nothing came easily for KC.
Before the turnaround SF had more total yards, were winning the turnover battle 2-1, and had a massive advantage in yards per play. KC only scored 10 points on their first five drives. On the other side of the ball SF had been stopped once by a turnover and once by poor clock management, but had scored on every other drive. There was little evidence to suggest the Chiefs could stop them. Everyone talks about the change on the Chiefs offensive side of the ball, but the 49es offense suddenly getting quickly stopped on two straight drives before failing on fourth down on the next drive was the real shocking change.
#38 by Will Allen // Feb 04, 2020 - 11:27am
This is why the criticism of Shanahan passing on 2nd and 3rd and five is so misplaced. Play action had been very successful up until then, the Chiefs had to be mindful of any Niner attempt to burn clock by running, and the plays were successful in getting the Niners best receiver wide open for easy throws. On 2nd down a Chief lineman made an excellent play to bat the pass down (not having looked at it closely again, I won't assign any responsibility to Garropolo) , and on third, with Kittle wide open for an easy, easy, throw, right in front of Garropolo, a mere 15 yards downfield, Garropolo throws to the wrong guy. A coach cannot call a game with the thought that his NFL qb won't execute a play that a qb in the MAC could pull off nine times out of 10. Guys have to make the easy, makeable, plays, in playoff games. Garropolo's foul up there was as egregious as the Saint cb who crapped the bed when trying to hit Diggs along the sideline two years ago, on the last play of the divisional game in Minneapolis.
#45 by Aaron Brooks G… // Feb 04, 2020 - 1:07pm
It is amusing the consider that each team faced a critical 2nd-5 where they wanted to burn clock.
SF, a team acclaimed for their running, passed and failed.
KC, a team acclaimed for their passing, ran and won.
I never saw it coming that SF couldn't run when their opponent knew they were going to run, and KC could.
#48 by Will Allen // Feb 04, 2020 - 2:27pm
Who knows? Maybe if the Niners run, they get 6 yards on 2nd and 5. But if Shanahan says to himself that a play action pass will result in Kittle beind wide open 15 yards downfield, right in front of Garropolo, for a toss that the qb at Kent State University would complete 90% of the time, and Shanahan turns out to be correct, but Garropolo screws it up, I can't really criticize Shanahan, any more than I would if he had called a run, and the play blew up when the guard whiffed on his block.
#40 by Eddo // Feb 04, 2020 - 11:43am
"Running more plays just means they were less efficient, and pretty much the same thing for time of possession."
While I agree with your first sentence (I don't think the Chiefs were more efficient, at least not until the final 10 minutes or so), this sentence is not really correct. Because of the concept of first downs, I would guess that efficiency actually correlates with more plays, when you have a large enough sample.
#52 by mrh // Feb 04, 2020 - 10:23pm
Me:
I don't buy the idea that the 49ers dominated thru 3.5 quarters... I'd say the 49ers were better - those turnovers counted - but not dominant.
You:
I don't think having more time of possession and running more plays indicates the Chiefs were playing better.
Then we agree: the 49ers were playing better. But I think you missed my point: those ToP and play numbers did not indicate "dominiance."
There were 121 games this year won by 10+ points. In only seven of them did the winning team have fewer than 27 minutes ToP (KC was on pace for over 33 minutes when down 10 after the 2nd INT). It is very unusual for a team to be winning that much, that late in a game, when the other team is controlling the clock to that degree.
Teams with 33+ minutes ToP were 99-27-1 this year, and out-scored their opponents by 8+ points. KC was playing generally winning football by this measure even at the point that Mahomes threw his 2nd INT - EXCEPT for that TO (and the fumble luck that was helping the Chiefs). Overall, of course, those TOs were huge and obviously a factor in SF's being ahead - and again, why I'd say the 49ers were playing better to that point, but not dominating.
What happened in the last 12 minutes was SF's offense got stopped by KC's defense, and Mahomes regressed to HIS mean of throwing TDs and not INTs. The Chiefs ended up with just over 33 minutes ToP in reality (vs. the pro-rated number above) and won by 10. That was pretty much an average outcome for a 2019 game.
#12 by theslothook // Feb 03, 2020 - 10:11pm
Can you imagine if Green Bay had won the Superbowl but came away w worse dvoa? A certain poster would be all over these boards. I'm glad that the Chiefs fans are happier with the win than demanding the dvoa bowl trophy along with it
#14 by Swilson1472 // Feb 04, 2020 - 2:02am
Chiefs fans that are also Royals fans know the power of cluster luck in winning championships. That Royals team also had their won their own 24-0 deficit equivalent (also against Houston) as well as ridiculous comeback wins in 3/4 wins they had in the World Series. Sometimes you just get lucky with stringing together good plays right next to each other at the end of a game.
#37 by Aaron Brooks G… // Feb 04, 2020 - 11:25am
The Royals stressed the Mets pitching with their running.
The 2014 and 2015 Royals were really good at taking extra bases and stealing bases, and even just the threat of this messed with pitchers. The won the 2014 WC game on the strength of 7 stolen bases.
#39 by Will Allen // Feb 04, 2020 - 11:34am
Really was a call back to 1970s baseball, which I think was more entertaining than the current paradigm of trading multiple strikeouts for a few homeruns. I understand the math of it, which makes the approach ruthlessly efficient, but it sure as hell isn't as fun to watch.
#26 by Jay Z // Feb 04, 2020 - 8:45am
But clustering didn't help the Chiefs at all. Their 3 big plays came on 3 separate drives. That they all came in the fourth quarter was incidental. Even after the 49ers fell behind, they still had plenty of time to respond and control the game. So I don't see what was "lucky" in what the Chiefs did.
#27 by Lost Ti-Cats Fan // Feb 04, 2020 - 8:49am
And sometimes you get unlucky when the other team strings a series of good plays together earlier in the game, and could be the only reason a late comeback is needed in the first place.
It's a four quarter game and KC was the better team over the full game, in my opinion. DVOA may have liked SF better in this game, but DVOA also liked KC better over the full season. DVOA typically prefers more consistent offensive production, whereas scoring quickly in bunches in between quarters of no scoring has been a hallmark of this year's Chiefs, making KC's full season DVOA lead even more telling.
#35 by Cythammer // Feb 04, 2020 - 11:00am
Do you have any evidence at all to suggest that the Chiefs have made a habit of "scoring quickly in bunches in between quarters of no scoring" besides your subjective impression? Secondly, do you have any reason to believe that such a pattern is anything other than meaningless, random noise?
#43 by Lost Ti-Cats Fan // Feb 04, 2020 - 12:24pm
The impression came from the KC games I'd watched this season, but I went back and looked at the box scores for their full season.
In 7 of their 16 games, KC scored 2+ TDs in a single quarter of the same game in which they scored 0 TDs in at least two other quarters. I don't, however, know how that compares to a typical NFL team.
My impression was certainly skewed by the games that I watched, the early Raiders game in particular. It sure looked to me like KC was of a higher probability than most teams to score in bunches, and in 14 of their 16 games this year they scored at least twice in a single quarter (though again, I'm not sure how that compares to an average NFL team).
I also don't have stats on how many deep plays KC successfully completed compared to most teams, but it seems to me like a lot of their multiple-scores per quarter are likely to have been a result of one or more short drives.
Maybe my impression's wrong and KC isn't really a dangerous offense because they have an above average ability to complete long passes and score quickly (and therefore in bunches), but I watch a lot of football, and from the games I saw that's how I'd characterize KC's O. If there are stats that say otherwise, I'm willing to stand corrected.
As for is that predictive, in loose terms, DVOA says not as predictive as short passing plays, in the sense that for the same amount of ground covered (50 yards, say) my understanding is that DVOA favours an offense that can consistently generate first downs and sustain drives over an offense that relies on the success of deep passes. Personally, though, I'm willing to accept the lack of data to support this, and say Mahomes-to-Hill has a higher probability of being replicated than a typical deep pass, and therefore DVOA may be understating the effectiveness of the 2019 KC offense.
#16 by ammek // Feb 04, 2020 - 2:14am
Why do you even read what that poster writes? It's gibberish.
If Green Bay had won the superbowl, it would feel very much like when the 2011 Giants or 2012 Ravens won it all. A bit surreal. A reminder that, for all the stats in the world, luck plays a bigger part in determining the outcome of individual games than we like to admit.
But Green Bay didn't win the super bowl or even play in it, so I don't know why I'm even talking about it. A win for stats! And the world is a slightly more just and comprehensible place, reassuring us for an instant, until we look out of the window again.
#33 by theslothook // Feb 04, 2020 - 10:37am
The Ravens also got a pick six on a pretty blatant pi penalty that went uncalled.
In any case, they were 13th in offensive dvoa and 19th in defensive DVOA, beating three superior teams. It was also powered by a string of games with Flacco playing way above his head.
#28 by Lost Ti-Cats Fan // Feb 04, 2020 - 8:53am
"A reminder that, for all the stats in the world, luck plays a bigger part in determining the outcome of individual games than we like to admit"
In fairness, good stats simple attempt to assess probabilities, not tell us what's going to happen in any given observation. If a 90% favourite fails to win, it doesn't mean the stats were wrong. If a 90% favourite loses 5 out of 6 games, then the stats may be off, but probabilities require a lot of observations, and sports championships give us very few. Intuitively, almost all of us struggle with that.
#34 by Lost Ti-Cats Fan // Feb 04, 2020 - 10:56am
There are some stats guys here that could give you a better answer, but without digging into my old stats books to refresh my memory about confidence levels, I can quickly tell you that a "fair" coin flipped 6 times will come up all heads 1.56% of the time and tails 1.56% of the time. So long streaks even between 50-50 teams aren't that uncommon. Although at that stage, the probability of the coin being "unfair" is obviously much higher.
Probably around 3 consecutive flips (12.5% odds for both tails and heads, i.e. 25% chance that one or the other gets flipped 3 times in a row) the odds favour the coin being "unfair" if you didn't have any external evidence to suggest that it should be fair,
#51 by Mountain Time … // Feb 04, 2020 - 5:40pm
My dad talks about a math teacher he had in the 1950s. They gave an assignment to the class to either flip a coin 100 times and record the results, or try to fool the teacher by making up results. The teacher could tell at a glance whether it was real or faked, because the real ones all had streaks of 5-6 heads or 5-6 tails in a row, and people making up the results just didn't know to include those streaks.
#55 by oaktoon // Feb 06, 2020 - 5:24am
And yes the Packers were, as everyone pretty much knew, including this certain poster and DVOA, not at the level of the 49ers.
But the basic point remains about this metric. It suggests that SF was, by a wide margin, the better team in the Super Bowl. They lost by 11 points and if you want to say that Williams' final touchdown only changed the final score and therefore has less siginficance, I would simply state that when he broke free the Niners still had timeouts and only trailed by 4 points. That play mattered.
So I am going to remain forever skeptical about a metric that concludes that the team that lost by 11 points in the biggest game of the season was "better". They were until they weren't. And the Chiefs whipped them good when it mattered most. Shanahan either didn't trust himself or didn't trust Jimmy G-- it matters not which and at what point in the game. DVOA might not be able to measure that element-- but we all saw it with our own eyes. Sure, the overthrow to Sanders might, if Garappolo had thrown that one pass accurately--have changed everything. But he didn't, and it didn't.
All I have ever said is that DVOA has flaws enough for people not to blindly follow it. Someday somebody will devise a better metric. SF under no definition of an athletic contest was the better team last Sunday. It was a perfect ending for an imperfect attempt this season to figure out what really happened in this league. The Patriots did not have a historic defense. The Niners were not a historically great team. Neither were the Ravens. Lamar Jackson came up small when his team needed him most. And unfortunately for my team, it failed to get home field advantage and thus had Aaron Rodgers try for a fourth time to get to a SB on the road-- my guess is he and they might be a bit better next year and fall into the same predicament. So it goes.
#58 by Eddo // Feb 07, 2020 - 5:02am
I mean, the team that had the higher DVOA heading into the game did win, so I'm not sure exactly how this is some soft of "gotcha" for Football Outsiders.
And if your point is that the team with the lower in-game VOA won, then I guess I see the logic... but what's the point of a metric that agrees with the game outcome 100% of the time? The point is that VOA saw the 49ers run more plays that are sustainable going forward, but in the small sample size of one game, it turned out the Chiefs' big plays came at the most opportune times to win.
#59 by Cheesehead_Canuck // Feb 09, 2020 - 8:56pm
Honest question. If you're so unconvinced by DVOA, why do you bother following this site? I think most people would acknowledge it has its flaws like any advanced metric does.
If SF was moderately better for 3 quarters and KC was significantly better for 1 quarter, it's not out of the realm of possibility that SF had better DVOA scores. And I'm pretty sure that Williams run really wasn't important. I don't think the 49ers were getting the ball back if he just falls down.
#60 by theslothook // Feb 10, 2020 - 1:49pm
Its hard to know where to begin with this post. Its simply breathtaking and highlights how thoroughly you misunderstand DVOA, and probability along with it.
And man, its even more shocking coming from a packers fan to read the phrase, " whipped them good when it mattered most." I wonder if Andy Reid was busy twiddling his thumbs in the third quarter because clearly that's just goof off time and its the 4th quarter where you bother showing up at all.
Taking your statement to its natural conclusions...Aaron Rodgers is not a very good qb since he cannot come through, "when it matters most".
#32 by beargoggles // Feb 04, 2020 - 10:33am
KC kind of flipped the script in their last 2 games by winning time of possession (I’m not certain about this in AFCC, but certainly they won the second half). Their own time consuming, run-infused drives really protected the Achilles’ Heel of their defense—having to tackle (Henry, Mostert, etc.) time after time in the 4th quarter. In the SB it probably really helped their 4th quarter pass rush as well.
#53 by beargoggles // Feb 05, 2020 - 12:37am
I was thinking of Titans and Niners games. Wasn’t sure about Titans in total, but pretty certain they hogged the ball in the third quarter after losing clearly this battle in the first half.
The Niner 3-and-outs (plus the endless supply of commercials) and Chief successful running might have killed the Niner 4th quarter running game among other things.